The inclusive-fitness model gives a powerful method for analyzing the sex-allocation patterns of social insects, both at the level of individual colonies and at the level of populations. Genetic relatednesses and reproductive values, both of which are essential in constructing the inclusive fitness, are relatively easy to estimate empirically. The reproductive success of males and females, however, can in many cases be difficult to estimate. This is particularly so when workers should be counted as a sexual investment, because colonies either recruit daughter queens as new reproductives or form new colonies by fission. My results suggest that we can predict equilibrium investment ratios in such cases, provided that workers are counted as an investment. When daughter colonies are formed by budding and they remain in close contact with the mother colony and with each other, the fitness consequences can be complicated, and the expected investment ratios depend on local resource competition or local resource enhancement. I suggest various alternative approaches for empirically testing the queen-worker conflict over sex allocation. The most promising approaches are those focusing on single species and comparing sex ratios produced by different colonies and populations. The inclusive- fitness model elaborated in this article can be applied to derive the expected allocation patterns, both as a function of the type of colony (number of reproductives in it) and as a function of the amount of resources the colony can use in sexual production.